Solving turbulence in a room

jipai · November 9, 2018, 17:59

Hi everyone,

I'm modelling the air flow in a room.

I successfully modelled the flow using simpleFoam without turbulence but when turning turbulence on, the solver diverges.

The turbulence fields are diverging starting at the edges between inlet/outlet and the air duct (wall). So I thought it is a stl problem according to the following post :

https://www.cfd-online.com/Forums/op...sh-salome.html

I tried different methods to optimise my mesh. For example, I used the tool surfaceMeshTriangulate to generate a waterproof mesh.

The solver still diverges almost immediately. So I'm muddled. Are the boundary conditions right configured ? Is there a problem in the settings of fvSchemes and fvSolution ?

Could any experienced user of snappyHexMesh and openFoam have a look and give my some hints please ? Is there a problem in the mesh generation ? In the stl files ? BCs ? Solvers settings ?

You can obtain my case following this link (8Mo) :

https://we.tl/t-KT2wBkQ89d

After downloading and extracting, just copy, paste and run :

./Allclean && cp -r 0.orig/ 0 && blockMesh | tee log.blockMesh && surfaceFeatureExtract | tee log.surfaceFeatureExtract && snappyHexMesh -overwrite | tee log.snappyHexMesh && checkMesh | tee log.checkMesh && potentialFoam | tee log.potentialFoam && simpleFoam | tee log.simpleFoam && simpleFoam -postProcess -func yPlus | tee log.simpleFoamYPlusRAS

Have a nice weekend all !

jipai

kerim · November 25, 2018, 17:01

Hi, jipai

Could you send your's case file once more time? previous was deleted.

jipai · November 25, 2018, 17:35

Hi Kerim,

Thank you for the reply

.

I sent you a new link :
https://we.tl/t-OAWBmEfvgd

Jipai

kerim · November 25, 2018, 20:23

Thanks a lot!

ssa_cfd · November 26, 2018, 05:16

Hi, I am also doing a airflow in room. Could you upload case file in google drive or github. I couldn't open the link.

or just post your files here..!

Thanks,
ssa.

jipai · November 26, 2018, 09:54

Hi ssa,

I shared the files on google drive. Please try this link :
https://drive.google.com/file/d/1c_l...ew?usp=sharing

The files takes 8Mo so cfd-online doesn't accept it

Have a nice day
jipai

ssa_cfd · November 26, 2018, 10:24

I just saw your files and here is the quick review.

You are using nutKwallfunction for k. use kqrwallfunction.
use wall function for nut.

From my opinion, try to use pimplefoam or pisofoam with RANS.

If you are running RANS, use upwind scheme for k,epsilon and omega. Don't use limitedlinear. If upwind works well, then change to linearupwind or other schemes.

jipai · November 26, 2018, 16:51

OK for kqRWallFunction

You're right about upwind scheme. It converges much better than limitedLinear (I actually manage to make the case converge with limitedLinear but with relaxation factors lowered to 0.5)

I will try to run with pimpleFoam or pisoFoam. I'm not yet familiar with transient similations with OpenFoam, but let's give it a try ^^

Many thanks for your hints
Regards

jipai

November 9, 2018, 17:59	Solving turbulence in a room	#1
jipai New Member Join Date: Oct 2015 Posts: 11 Rep Power: 10	Hi everyone, I'm modelling the air flow in a room. I successfully modelled the flow using simpleFoam without turbulence but when turning turbulence on, the solver diverges. The turbulence fields are diverging starting at the edges between inlet/outlet and the air duct (wall). So I thought it is a stl problem according to the following post : https://www.cfd-online.com/Forums/op...sh-salome.html I tried different methods to optimise my mesh. For example, I used the tool surfaceMeshTriangulate to generate a waterproof mesh. The solver still diverges almost immediately. So I'm muddled. Are the boundary conditions right configured ? Is there a problem in the settings of fvSchemes and fvSolution ? Could any experienced user of snappyHexMesh and openFoam have a look and give my some hints please ? Is there a problem in the mesh generation ? In the stl files ? BCs ? Solvers settings ? You can obtain my case following this link (8Mo) : https://we.tl/t-KT2wBkQ89d After downloading and extracting, just copy, paste and run : ./Allclean && cp -r 0.orig/ 0 && blockMesh \| tee log.blockMesh && surfaceFeatureExtract \| tee log.surfaceFeatureExtract && snappyHexMesh -overwrite \| tee log.snappyHexMesh && checkMesh \| tee log.checkMesh && potentialFoam \| tee log.potentialFoam && simpleFoam \| tee log.simpleFoam && simpleFoam -postProcess -func yPlus \| tee log.simpleFoamYPlusRAS Have a nice weekend all ! jipai Last edited by jipai; November 11, 2018 at 09:17.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
conjugate heat transfer in OpenFOAM	skuznet	OpenFOAM Running, Solving & CFD	99	March 16, 2017 05:07
Cannot run the code properly: very large time step continuity error	crst15	OpenFOAM Running, Solving & CFD	9	December 14, 2014 18:17
calculation stops after few time steps	sivakumar	OpenFOAM Running, Solving & CFD	7	March 17, 2013 06:37
Error while running rhoPisoFoam..	nileshjrane	OpenFOAM Running, Solving & CFD	8	August 26, 2010 12:50
IcoFoam parallel woes	msrinath80	OpenFOAM Running, Solving & CFD	9	July 22, 2007 02:58

November 25, 2018, 17:01		#2
kerim Senior Member abdikerim kurbanaliev Join Date: Jun 2010 Location: Kyrgyzstan, Osh Posts: 120 Rep Power: 15	Hi, jipai Could you send your's case file once more time? previous was deleted.

November 25, 2018, 17:35		#3
jipai New Member Join Date: Oct 2015 Posts: 11 Rep Power: 10	Hi Kerim, Thank you for the reply . I sent you a new link : https://we.tl/t-OAWBmEfvgd Jipai

November 25, 2018, 20:23		#4
kerim Senior Member abdikerim kurbanaliev Join Date: Jun 2010 Location: Kyrgyzstan, Osh Posts: 120 Rep Power: 15	Thanks a lot!

November 26, 2018, 05:16		#5
ssa_cfd Member ssa Join Date: Sep 2018 Posts: 93 Rep Power: 7	Hi, I am also doing a airflow in room. Could you upload case file in google drive or github. I couldn't open the link. or just post your files here..! Thanks, ssa.

November 26, 2018, 09:54		#6
jipai New Member Join Date: Oct 2015 Posts: 11 Rep Power: 10	Hi ssa, I shared the files on google drive. Please try this link : https://drive.google.com/file/d/1c_l...ew?usp=sharing The files takes 8Mo so cfd-online doesn't accept it Have a nice day jipai

November 26, 2018, 10:24		#7
ssa_cfd Member ssa Join Date: Sep 2018 Posts: 93 Rep Power: 7	I just saw your files and here is the quick review. You are using nutKwallfunction for k. use kqrwallfunction. use wall function for nut. From my opinion, try to use pimplefoam or pisofoam with RANS. If you are running RANS, use upwind scheme for k,epsilon and omega. Don't use limitedlinear. If upwind works well, then change to linearupwind or other schemes.

November 26, 2018, 16:51		#8
jipai New Member Join Date: Oct 2015 Posts: 11 Rep Power: 10	OK for kqRWallFunction You're right about upwind scheme. It converges much better than limitedLinear (I actually manage to make the case converge with limitedLinear but with relaxation factors lowered to 0.5) I will try to run with pimpleFoam or pisoFoam. I'm not yet familiar with transient similations with OpenFoam, but let's give it a try ^^ Many thanks for your hints Regards jipai